• Ei tuloksia

Novel Methods for Assessing the Protection Impacts of Distributed Generation in Distribution Network Planning

N/A
N/A
Info
Lataa
Protected

Academic year: 2023

Jaa "Novel Methods for Assessing the Protection Impacts of Distributed Generation in Distribution Network Planning"

Copied!
90
0
0

Kokoteksti

(1)

Kari Mäki

Novel Methods for Assessing the Protection Impacts of Distributed Generation in Distribution Network Planning

Tampere 2007

(2)

Tampere University of Technology. Publication 701

Kari Mäki

Novel Methods for Assessing the Protection Impacts of Distributed Generation in Distribution Network Planning

Thesis for the degree of Doctor of Technology to be presented with due permission for public examination and criticism in Sähkötalo Building, Auditorium S1, at Tampere University of Technology, on the 5th of December 2007, at 12 noon.

Tampereen teknillinen yliopisto - Tampere University of Technology Tampere 2007

(3)

ISBN 978-952-15-1884-3 (printed) ISBN 978-952-15-2737-1 (PDF) ISSN 1459-2045

(4)

ABSTRACT

Recent developments in energy policies and prices have directed an increasing amount of interest at exploiting small energy resources. This small-scale power generation evidently needs to be connected to the present power distribution system with simple manners in order to be economically competitive. As the distribution system has initially been built for simple one-way power delivery, interconnecting small generators requires new thinking and new methods for operating the network. Small-scale generators located on the distribution level are generally referred to as distributed generation (DG).

There are several clear consequences for locating DG in the distribution network that need to be taken into account. Probably the most critical concern is the operation of network protection. This can be stated as protection malfunctions can result in safety hazards. The presence of DG affects for instance short-circuit current amplitudes, which can further disturb the operation of feeder protection.

DG may also result in failing reclosings or situations in which the DG unit maintains the voltage alone in the network. These situations must be avoided.

This thesis focuses on the network protection impacts of DG. It aims to provide new tools and methods for including the impacts of DG correctly in network planning methods. This can be achieved by bringing the research observations towards the practical level of network planning. To help form a precise image on the subject, the typical DG-related protection problems are described and analysed. Simulation tools have been used to study the phenomena. A perspective of distribution network planning is maintained throughout the thesis. The information systems applied for planning purposes are also considered.

Development needs created by DG are especially focused on.

The most important observations of this thesis relate to the contradictions between protection selectivity and sensitivity; to the problematic nature of differentiating between faults that require rapid actions and other disturbances that should not result in any action. At the present situation, it may be possible to trip the DG units too sensitively in the name of network safety. However, as the DG becomes more significant for the power system, more coordination will be needed in the future. An efficient coordination of protection devices during all possible situations is beneficial to all parties.

The thesis shows the significance of performing certain studies during the interconnection process of a new DG unit. The sequence of the studies performed can also be essential depending on the case. The thesis proposes a procedure for

(5)

gathering the studies as one entity. A new approach for presenting protection requirements for the new DG unit is also presented. This method is intended especially for facilitating the dialogue between network utility and power producer. A new method for extending the short-circuit calculation of a typical planning system is also defined. This extension enables more accurate results.

The ideas presented for integrating the DG impacts in planning systems and methods form the most important contribution of this thesis. The knowledge gathered and case studies conducted should also be useful for the interest groups of DG.

According to many forecasts, the amount of DG will rapidly increase in the near future. At the same time, managing the DG impacts will become an essential question for the network utilities. It can be seen that the planning process for a new DG unit interconnection can become a part of daily network planning activities. Thereby it needs to be handled efficiently.

(6)

PREFACE

This work has been carried out during the years 2003-2007 in the Institute of Power Engineering at Tampere University of Technology (TUT). The supervisor of this thesis has been Professor Pertti Järventausta. I would like to express my gratitude to him for his support and guidance. I am also grateful to Sami Repo, Dr.

Tech., for good advice and support.

I would like to thank the personnel of the Institute of Power Engineering at TUT for a comfortable working environment. I especially want to thank my colleagues Anna Kulmala, M.Sc., and Jussi Antikainen, M.Sc., for collaboration and discussions. I would also like to thank Hannu Laaksonen, M.Sc., nowadays with University of Vaasa, for collaboration during the early stages of the research. A special word of thanks must be addressed to Merja Teimonen, institute secretary, for great arrangements.

The research behind this thesis has been conducted under a few research projects in cooperation with the The Finnish Funding Agency for Technology and Innovation (Tekes), Fortum Sähkösiirto Oy, Vattenfall Verkko Oy, Rovakaira Oy, Kemijoki Oy, Fingrid Oyj, ABB Oy, Vamp Oy, Wärtsilä Finland Oy, Nokian Capacitors Oy, Cybersoft Oy, MX Electrix Oy, Koillis-Satakunnan Sähkö Oy, Tampereen Sähkölaitos and the Finnish Energy Industries Federation (Finergy). I would like to thank all project parties for a fruitful and interesting collaboration.

The financial support provided by Tekniikan Edistämissäätiö and Walter Ahlström Foundation is gratefully acknowledged.

I would like to thank my parents and my sister Sirpa for all the support and encouragement. I am also grateful to Sirpa for checking and improving my English.

I want to thank Tiina, Roosa and Nero – my family – simply for everything.

Finally, this thesis is dedicated to the loving memory of Jedi, a great small dog who supported and encouraged me more than most people can ever realise.

You would have deserved so much more.

Kangasala, November 2007

Kari Mäki

(7)

TABLE OF CONTENTS

Abstract... i

Preface ... iii

Table of contents... iv

Publications... vi

List of symbols and notations ... vii

1 Introduction...1

1.1 Role of distributed generation...2

1.1.1 Motivation for DG propagation ...2

1.1.2 Interest groups of DG ...4

1.1.3 Definition of DG ...5

1.1.4 Present situation of DG protection...6

1.2 Contribution and context of the thesis ...8

1.2.1 Motivation and objectives...8

1.2.2 Outlining of the thesis...9

1.2.3 Publications and evolution of work ...9

1.2.4 Structure of the thesis ...11

1.3 Terminology...12

2 Impacts of DG on distribution network protection ...16

2.1 Protection impacts...17

2.1.1 Impact of generator type ...19

2.1.2 Impact on short-circuit currents ...20

2.1.3 Sensitivity problems ...21

2.1.4 Selectivity problems ...24

2.1.5 Failed reclosing...26

2.1.6 Loss-of-mains detection...28

2.1.7 Earth fault detection...30

2.2 Aspects of network planning ...31

2.3 Research activities on DG protection impacts ...33

(8)

3 Network planning and calculation tools ...38

3.1 Network information system as a practical-level planning tool ...38

3.2 Dynamical simulation tools for research purposes ...41

3.3 Real-time systems for testing and simulating purposes...42

3.4 Development needs posed by DG in daily network planning ...43

4 Studies performed and methods developed...47

4.1 Studies performed ...48

4.1.1 Overview of simulation results ...49

4.2 General protection planning procedure...54

4.3 Protection requirement graph...55

4.4 Calculation extension for network information systems ...56

4.5 Automating the methods in information systems ...57

4.6 Work to be done...57

5 Summary and discussion...59

6 References...62

(9)

PUBLICATIONS

The thesis consists of the following publications:

1. Mäki K., Järventausta P., Repo S., Protection Issues in Planning of Distribution Network including Dispersed Generation. CIGRE Symposium: Power Systems with Dispersed Generation, Athens, Greece, April 13-16, 2005, paper 1P-05, 8 p.

2. Mäki K., Repo S., Järventausta P., Protection Coordination to meet the Requirements of Blinding Problems caused by Distributed Generation.

WSEAS Transactions on Circuits and Systems, Vol. 4, Issue 7, July 2005, pp. 674-683

3. Mäki K., Repo S., Järventausta P., Network Protection Impacts of Distributed Generation – A Case Study on Wind Power Integration.

Nordic Wind Power Conference 2006, Espoo, Finland, May 22-23, 2006, 6 p.

4. Mäki K., Kulmala A., Repo S., Järventausta P., Problems related to Islanding Protection of Distributed Generation in Distribution Network. Power Tech 2007 conference, Lausanne, Switzerland, July 1- 5, 2007, 6 p.

5. Mäki K., Repo S., Järventausta P., Impacts of Distributed Generation on Earth Fault Protection in Systems with Isolated Neutral. CIRED International Conference on Electricity Distribution 2007, Vienna, Austria, May 21-24, 2007, paper 0107, 4 p.

6. Mäki K., Kulmala A., Repo S., Järventausta P., Studies on Grid Impacts of Distributed Generation in a Combined Real-Time Simulation Environment. IPST International Conference on Power Systems Transients, Lyon, France, June 4-7, 2007, 6 p.

7. Mäki K., Repo S., Järventausta P., General Procedure of Protection Planning for Installation of Distributed Generation in Distribution Network. International Journal of Distributed Energy Resources, Vol.

2, No. 1, January-March 2006, pp. 1-23

8. Mäki K., Repo S., Järventausta P., Protection Requirement Graph for Interconnection of Distributed Generation on Distribution Level.

International Journal of Global Energy Issues, Vol. 28, No. 1, 2007, pp. 47-64

9. Mäki K., Repo S., Järventausta P., New Methods for Studying the Protection Impacts of Distributed Generation in Network Planning Systems. 15th International Conference on Power System Protection, Bled, Slovenia, September 6-8, 2006, pp.128-133

(10)

LIST OF SYMBOLS AND NOTATIONS

CCT Critical clearing time CFP Common feed point CHP Combined heat and power DC Direct current

DG Distributed generation DLC Distribution line carrier

DMS Distribution management system DNO Distribution network operator FCL Fault current limiter

IPP Independent power producer

LV Low voltage

MV Medium voltage

NDZ Non-detection zone

NIS Network information system PCC Point of common coupling RES Renewable energy source rms Root mean square

ROCOF Rate of change of frequency RTDS Real Time Digital Simulator

SCADA Supervisory control and data acquisition TCM Time coordination method

THD Total harmonic distortion

TUT Tampere University of Technology VS Vector shift

(11)
(12)

1 INTRODUCTION

Local power generation is not a new phenomenon at all. Actually, the initial forms of power systems comprised of local generation units, for instance hydro power, which were feeding local loads. As the amount of installations requiring electricity increased, a need for a more reliable and economical power delivery arose. Further, this led to a need for a wider power delivery system. As entire power networks were constructed, the power generation was concentrated in larger, centralized power plants. This centralized development absolved the customers of maintaining their own generation. Furthermore, larger generating unit sizes improved overall efficiencies.

The present power systems are still mainly based on centralized power generation.

This has been considered as the sole settlement, which has led to extending the electrical network in all areas with demand for electrical energy. Practically, this has resulted for instance in long transmission and distribution lines on rural areas with low demand. In the present situation the distribution network is truly extensive in the Nordic countries, which has actually led to concerns of maintaining the furthest parts of the networks. Changes in community structure have a significant impact on this issue; due to the migration towards cities, fewer and fewer customers remain in rural areas. Evidently, this results in reduced distribution incomes but does not absolve the distribution network operator (DNO) from its responsibilities in maintaining the network and assuring proper power quality.

Recent developments have once again steered the public interest towards small- scale local generation of power and electricity. Environmental reasons and pursue for improved energy efficiency are typical reasons for this reversion. The centralized evolution described above is actually an enabling factor regarding the revival of local generation. This is because the density of the distribution network offers great possibilities for exploiting even small energy resources without major investments to the network.

(13)

1.1 Role of distributed generation

For a distribution network, the increase of distributed generation (DG) represents a need for reshaping the overall philosophy for planning and controlling the network. The situation may be problematic as the parties related to DG have different objectives and motives. In addition to the network company and the power producer there are many other parties that are very interested in DG at the moment. Actually there are lot of expectations and hopes regarding DG and future power production.

1.1.1 Motivation for DG propagation

There are certain drivers promoting a more wide-scale usage of DG in distribution networks. One major factor is that the some techniques suitable for DG units offer environmentally sound options for power production. This relates directly to the need for reducing greenhouse gases, to the emission rights and to the Kyoto protocol. DG does not automatically mean clean energy production, but it provides a possibility for utilizing local energy resources that would not be economically viable in greater scales. Thus DG is often associated with wind power, solar power, small-scale hydro power, bio-mass based production etc.

Another significant driver is the increasing price of electricity. As electricity – as well as other forms of energy – becomes more expensive, small units utilizing local resources become more attractive. Typically, the source of primary energy is local and is not subject to rapid price variations or to unexpected availability problems of raw materials. Other economical drivers include short lead times of building processes and modularity of small units. In the case of multiple units (for instance a wind park) it may be possible to use the first unit for energy production while the other ones are still being built. Together with short lead times, this enables profit for the invested capital already in the early stage of the project.

Further, it is often possible to add more DG units later to the site without significant investments to the infrastructure. The earlier listed environmental benefits of DG are often transformed to economical benefits through financial support, emission taxes, etc. The society plays an important role in this sense.

Power production, network operation and energy retailing are separated in the present legislation with deregulated electricity markets. This means that parties investing in DG units are independent power producers (IPPs) who are usually not related to the local DNO. From the IPP’s point of view, the economical issues mentioned are the most important drivers. For the DNO, the possibilities of increasing network reliability and improving voltage profile at the network are the most desirable outcomes of DG.

(14)

DNO has typically no possibility of controlling the DG unit and, on the other hand, DNO can not act as a power producer on the liberalised energy market.

Thereby DG units for increased reliability mean in many cases movable or immovable stand-by power units. These units do not run in parallel with the public network, which explicitly differentiates them from other forms of DG. The difference between stand-by power and independently running DG is essential, as it is explained later in this chapter. This thesis focuses on independent DG units. It can be said, that these units do not actually contribute to the network’s reliability as long as there are no contracts between the DNO and the IPP on running the units according to the network’s needs. On the contrary, they may reduce the reliability by increasing the number of interruptions and faults as it can be seen later. They may also result in an increased number of complaints on power quality as normal network level faults often cause harm to the IPP.

At a more general level, including both economical and environmental aspects, it can be stated that power should always be produced where it can be done efficiently and economically. DG can help in achieving this goal by providing relatively dense distribution networks for power production. The idea of DG is based on a simple and inexpensive connection to the network, which evidently leads to small-scale units, which is further highly compatible with the transfer capacity of distribution networks. In the cases in which DG generates significant portions of the local consumption, it can reduce the need for a reliable distribution network. At its extreme, this can mean a local DG unit which feeds the load of the customer, hence leaving the network connection as backup only. On the other hand, if DG is apart from loading and the power is thereby transmitted longer distances, it further emphasizes the need for a reliable distribution network.

Combined heat and power (CHP) production is an excellent example of what DG typically offers to the power system. It is practically always based on a customer or customer group requiring district heating, process steam, cooling, clean water, etc. This requirement is satisfied for instance by burning fossil fuels or biomass.

Adding a turbine and a generator to this heating plant offers a cost-effective way of generating electrical power from the energy that would otherwise be lost. At the same time, the overall efficiency of the plant is increased significantly. On the other hand, the electrical power production is dominated by the needs of heating or other primary purpose of the unit. Only a limited possibility of controlling the electrical power produced can be achieved for instance by using heat accumulators. This can be made to optimize the income from the electric power produced. However, from the DNO’s point of view, the production is more or less intermittent. CHP fulfils thus the typical characteristics of DG; the energy resource is locally available, relatively small investments are needed and the main issue lies in integrating it efficiently with the electrical network. As a drawback, it

(15)

must be accepted, that the generation can not be controlled by the DNO, for instance according to the loading of the network. These issues are highly linked with the definition of DG, which is covered later in this chapter.

As an addition to the CHP example, it must be noted that CHP is almost an ideal form of DG as it is, in addition to the mentioned characteristics, often located near to the customers and the electrical power consumption tends to follow the outdoor temperature at least in the Nordic countries. For instance wind power shares the other basic DG characteristics, but is typically located further from the customers and is less likely to follow loading of the distribution network.

1.1.2 Interest groups of DG

As it has already come up, there are several parties that relate closely to DG. The typical interest groups of DG are shown in figure 1.1. The most important player is the power producer, IPP, who has the initial motive to build the DG unit and to connect it to the distribution network. DNO has an essential role in providing the distribution network and assuring the safety and functionality of it, also after connecting DG. Safety of the network is typically assured by defining suitable requirements for the DG interconnection. Safety level has also been assured earlier with type and commissioning tests performed for each unit type. The manufacturer or supplier of the DG unit also has an important role. It is common that the supplier interacts directly with the DNO during the DG planning process.

Special consultants can also be used to manage the planning and communication between the parties.

The society is very interested in DG progress and is thus an important interest group. The most desired outcome from this point of view would be a rapid increase in environmentally friendly energy production – in other words, reduction of pollution and greenhouse gases. The society is also dealing with DG in the forms of legislation, environmental aspects, land ownership and allocation, construction permissions, taxation, electricity market regulation and so on.

DG also offers new possibilities for different ancillary service providers. One important form is likely to be measuring services. DG always requires some basic measurements, whereas additional measurements could be very useful for both IPP and DNO. Power quality monitoring including analysis services is one form of ancillary services that relates closely to measurement services. Consultant services focusing on the studies needed for DG interconnection could find customers among both DNOs and IPPs. Service providers operating for instance in the areas of environmental planning or construction can also specialize in different forms of DG. Experts on taxation and subsidies are also needed when

(16)

considering the financing of a new DG unit. Also the financiers and investors form an important interest group. In the case of DG units running with biofuels, new possibilities have already been found in biomass harvesting and processing.

Distributed generation

Research and development Distribution network

operator Power

producer

Manufacturer

Financiers and investors

Society

- Environmental possibilities - Energy supply aspects

- Permissions and land allocation

- Legislation - Taxation - Market regulation

Service providers

- Consultants - Measurements

- Construction - Fuel service

Figure 1.1. The interest groups of DG.

1.1.3 Definition of DG

The definition of DG is by no means an unambiguous issue. The definition can be made according to many different factors. The definition applied in this thesis is clearly based on the distribution network aspect, emphasizing thus the network connection instead of the primary energy resource. The definition stated hereafter is applied in this thesis and might differ from definitions given in other sources.

DG is often defined through the scale of the generator units or the type of the primary energy resource. It is very common to define a certain power or voltage level limit, below which generation is defined as DG. It is also very common to list typical energy resources, for instance wind power, as forms of DG. However, as wind power is increasingly built in big wind parks that are connected to the transmission level, they can not be automatically defined as DG.

Instead of the type of the generating unit, DG can be defined as a way of locating generation in the power system. This approach is applied in this thesis. DG is located in the network according to the primary energy resource and operates

(17)

independently regardless of the distribution network’s state when the network is available. DG is connected to a low or medium voltage network. DG is running in parallel with the network, which unambiguously separates it from the stand-by power units. DG may be located in the public network or in the customer’s own network. It may be connected directly or via a unit generator transformer. DG is not controlled by DNO and the location of new units can not be optimized on the grounds of network’s characteristics. The timing of the DG construction can neither be controlled by the DNO. Neither the size of the unit nor the primary energy resource is defined. In other words, the location of generation is much more essential than the actual energy form.

It is also evident, that certain types of energy production are more suitable for DG applications than others. The transfer capacity of low and medium voltage distribution networks limits the generated power. As described above, the concept of DG is based on local energy resources. Thereby the typical DG resources include for instance wind, solar, hydro, CHP, diesel engines etc. However, the resource is not used as a definition in this thesis.

According to the principles of liberalised energy markets, DG is assumed to be owned and driven by the IPP only. DNO can not own DG units and, especially, it can not locate them according to the network’s needs. Thereby the methods for optimizing the location of DG in the network are not considered in this thesis. The location of DG is assumed to be defined by other factors as explained later. The question of optimizing the DG location is relevant in areas where DNO can act as a power producer as well. In this case, building a new DG unit may even be used as an option to building new lines or reinforcing existing ones. Thus the planning tasks would change significantly, although the technical issues remain the same.

Condensed, two essential points must be underlined. In this thesis:

• DG does not equal to renewable energy resources.

• DG refers more to the way of locating generation in the distribution network than to certain forms of power generation.

1.1.4 Present situation of DG protection

At the moment the rapid increase in the amount of installed DG units seems obvious. At the same time, the awareness of the grid impacts of DG is increasing.

So far, most of the possible protection problems have not been checked during DG installations. On the other hand, most of the installations have never faced such problems as the problems usually require worst-case circumstances to occur.

Presently, DNOs are facing problems especially with their first DG installations or in networks with high DG penetration levels. As the protection consequences

(18)

often relate to worst-case faults only, they may remain hidden in the network as long as such situation occurs. Thereby the need for extensive protection analysis during the DG interconnection process must be underlined.

As DG has become more significant for the whole power system, the fault ride- through requirements have been introduced in order to maintain stability during different disturbances. From the protection perspective, these requirements may be problematic as they are driving the situation towards less sensitive protection in the point of common coupling (PCC). On the other hand, it is evident that it is beneficial for all parties to keep the DG unit in the network when there is no need for tripping it. Present methods lead easily to compromising between selectivity and sensitivity. Differentiating between faults requiring actions and other disturbances will be the essential question in the future.

At the moment, a significant part of the new DG units are equipped with power electronic converters. A converter usually defines the operation of the DG unit during faults despite the source of primary energy. Converters are generally less problematic than the traditional generators, as they usually do not contribute to system short-circuits in as significant way as traditional generators do. The short- circuit contribution of a converter is often only a momentary peak, although it is possible to design a converter for feeding prolonged short-circuit currents as well.

A converter is a complex component as its behaviour is strongly dependent on the design and is relatively unknown from the power system’s point of view.

It can be seen, that the PCC protection is at the moment moving towards the converter instead of dedicated relays. The converter could be able to perform many of the same actions as the relay. The present converter installations already contain unnecessary overlapping, for instance in the form of measurements. Thus economical advantages are observable, but from the DNO’s point of view the reliability of such configuration is difficult to evaluate. Using a converter for fault detection and tripping would naturally require commissioning tests similar to the ones performed on new relay types. However, the general accessibility and, for instance, the possibility of modifying the operation values of a protection system integrated to the converter could be considered problematic from the DNO’s point of view.

Generally, the situation seems more or less confusing with different requirements, grid codes, new techniques, unsolved problems (i.e. loss-of-mains) and so on. The fault ride-through requirements are complicating the situation. The practices applied in different countries vary quite a lot [1]. In the practical DG planning actions, uniform procedures may be rare even within one DNO, not to mention the whole distribution network sector. Development is definitely needed.

(19)

1.2 Contribution and context of the thesis

The purpose of this thesis is investigating the network protection impacts of DG and developing methods for assessing them as a part of the practical-level network planning activities. The objectives of the research are focused on usage of advanced network information systems. The point of view applied is clearly on the power system’s and DNO’s side.

1.2.1 Motivation and objectives

The studies performed have been especially motivated by two facts:

• It can be seen that managing the protection impacts of DG will play an essential role in the near future considering the propagation of DG. The minimum requirements will be to assure that the distribution network is used safely and that the protection will not form obstacles before the progress of DG.

• The knowledge on DG interaction among DNOs is more or less superficial.

This is natural as the DG interconnections are individual cases at the moment. Similarly, the present planning systems do not support the DG studies as they could do. DNOs will be facing the DG interconnection studies more and more often in the near future. Efficient methods and tools will be needed.

Due to the above mentioned facts, the protection impacts of DG units must be included in the daily network planning and operation activities instead of performing only case-specific studies.

The aim of the thesis can be condensed as follows:

• To provide information and knowledge on coordinating the protection system in the presence of DG. This means correct co-operation of feeder relays and DG protection during different faults and disturbances.

Nuisance tripping avoidance is an essential part of this objective.

• To provide methods for developing the present network planning systems to better support the planning process. This requires for instance development of generator modelling in present systems.

• To introduce practical-level planning rules and principles. This means bringing the research observations closer to the network planning activities.

(20)

1.2.2 Outlining of the thesis

As a definition of this thesis, an exclusionary border could be drawn in the connection point of the DG unit. This thesis studies the power system side and does not commit to the primary energy resource nor the control system applied in the DG unit. The internal protection and control systems of the DG unit are not considered either. On the other hand, the protection at the DG connection point is of great interest. Further, the thesis does not intend to propose new protection techniques or to offer protection relay development. On the contrary, it intends to cover the possibilities of applying present commonly used techniques in the presence of DG.

The studies presented are performed primarily in a typical Nordic distribution network. This means a medium voltage (MV), symmetrically loaded three-phase radial system with isolated or compensated neutral. The studies presented are performed mainly in 20 kV distribution network. The studied networks consist mainly of overhead lines and are equipped with feeder-dedicated protection relays located at the primary substation. These relays are usually overcurrent relays operated with definite time characteristics. Time-dependent operation is, however, included in the developed methods. No additional relays, breakers or fuses are located along the feeders, but otherwise the grade of automation is high with a lot of remotely controlled switches, measurements etc. The feeders studied include both generation and loading, which has been the case so far in most installations in Finland. Dedicated feeders for connecting the DG unit to the substation have thus not been considered in detail. Technically, the observations made apply to these dedicated feeders as well, but they are far easier to handle as the feeder protection can be adjusted according to the requirements of the DG unit.

Many of the observations made can be directly generalized to other network types, whereas some (especially earth fault issues) may depend strongly on the type of the network.

1.2.3 Publications and evolution of work

The research work behind this thesis has evolved from determining the protection impacts of DG to offering general planning principles through many case-studies, simulations and calculations. The publications on these subjects are included in this thesis. Figure 1.2 illustrates the evolution of the research work.

(21)

Figure 1.2. Position of publications in the research context.

The thesis includes nine international publications:

• Publication 1 discusses the general planning perspective of the issue. Some studies comparing dynamical simulation tools with practical network planning systems are performed.

• Publication 2 covers the protection blinding phenomenon. Theoretical background of blinding is discussed. Principles for assessing the intensity of blinding are presented. Results for example studies are presented.

• Publication 3 focuses on a case study of relatively small-scale wind power units in a realistic network. The main emphasis is on voltage dips during different faults and on the possibility of nuisance trippings of the units.

• Publication 4 focuses on islanding detection problems. Reasoning as well as fundamental problems of islanding are explained. The operation of different protection methods during difficult islanding conditions is studied. The need for coordination to avoid unnecessary trippings is highlighted again. However, new protection methods are not pursued.

• Publication 5 studies the earth fault situations. Especially the needs and possibilities for detecting the system earth faults at the DG unit’s connection point are discussed. Faults on the adjacent feeders and their potential for nuisance trippings are also studied. Specific new protection methods are not proposed.

(22)

• Publication 6 presents the integrated research environment for power system studies. As an example, the system is used for studying the basic protection impacts of DG. Typical phenomena are observed, although the studied case is not very problematic.

• Publication 7 introduces the procedure for protection planning during DG interconnection process. It seeks to include all necessary studies in correct sequence. The iterative nature of the studies needed is an essential characteristic of the procedure. The procedure is linked with the methods presented in publications 8 and 9.

• Publication 8 proposes a novel method of presenting the protection requirements for a new DG unit. The graphical approach presented is based on calculations of network information systems. The method has been designed for providing an output of the procedure of publication 7.

• Publication 9 presents an idea for expanding the short-circuit calculation of typical network planning systems. The method is based on iterative looping of typical calculations. Comparisons are made between a typical information system calculation and the expanded one. The ideas presented are meant to support the studies needed in the methods presented in publications 7 and 8.

The author of this thesis is the corresponding author of all nine original publications. The author has contributed in all publications in the form of literature surveys, calculations, modelling, simulations, analysis and reporting. All results presented are based on author’s original work. The planning methods developed are totally based on author’s original ideas and development work.

All publications have been written in collaboration with Dr.Tech Sami Repo, whose contribution has been mainly in comments supervising the research. Some of the simulation models applied in the studies of publication 3 were based on HElib Model Library created by VTT and University of Vaasa. Simulations using the combination of Real Time Digital Simulator (RTDS) and dSPACE that are presented in publication 6 have been performed in collaboration with M.Sc. Anna Kulmala and personnel of the Institute of Power Electronics at TUT. Professor Pertti Järventausta contributed to all the publications by supervising the research.

1.2.4 Structure of the thesis

Chapter 2 presents the network protection impacts related to the integration of DG units. After introducing the impacts from the DNO’s point of view, chapter 2 discusses the planning aspects focusing on how the impacts presented should be taken into account in practical network planning. Chapter 3 presents different tools for studying the phenomena of chapter 2. Typical practical-level planning

(23)

tools as well as purely research-intended tools are introduced. As the most important part, possibilities of importing new features to the planning systems according to experiences in research environments are discussed. Chapter 4 presents the methods developed during research actions. An overview of performed simulations is given. Future work is also ideated. Chapter 5 draws a summary and concludes the contents of this thesis.

1.3 Terminology

Most of the technical terms used in this thesis comply with the IEC Vocabulary.

Terms that are not defined in the vocabulary are defined in this chapter. Some of the definitions may differ from definitions given in other sources, but they are applied throughout this thesis.

Anti-islanding protection

The power system can be used as an island when a part of the network remains energized by local generation without connection to the main system. The transition to the island is called islanding. The term anti-islanding protection refers to the detection and tripping of unintended islanding. However, the term is somewhat misleading as the island is to be prevented, not the islanding itself.

Instead, islanding has to be detected in order to prevent sustained islands. More precise terms loss-of-mains protection and islanding detection have been used throughout the introductory part, whereas the term anti-islanding has also been used in the publications included.

Blinding ratio

Blinding ratio is a term defined by the author in order to illustrate the intensity of the blinding effect. The ratio is calculated according to the proportion of impedances between the common feed point and the fault point, DG unit and the feeding substation, respectively. The calculation method is presented in publication 2.

Block transformer

The term block transformer refers to the transformer used to connect the DG unit to the medium-voltage network. This transformer is used only for transforming the voltage of the unit to the network voltage; hence no load demand is fed by the transformer. The more precise term unit generator transformer is used throughout the introductory part, whereas the term block transformer has been used in publications included.

(24)

Common feed point

Common feed point is a term defined by the author to illustrate the theory of the blinding effect. Common feed point is the point closest to the fault that is yet fed in parallel by the DG unit and the feeding substation. Thus the location of the common feed point is unique for each fault studied. The most important purpose of the concept of common feed point is to clarify the calculation of blinding impact. The calculation can be initiated by finding the common feed point for the fault studied.

Directional relay

Directional relay is a feeder relay, which is able to recognize the direction of the current and to operate differently according to this direction. Practically, the operation of relay is allowed for one current direction only.

Distribution management system

Distribution management system is an information system for on-line monitoring of the medium-voltage distribution network. Network components can be measured and controlled via SCADA system. The system includes load data and it is hence used for distribution network state estimation. Other features for supporting the network operations such as fault location, switching planning and network optimization are also included. Among others, these features differentiate the distribution management system from a SCADA system.

Fast or instantaneous autoreclosing

Terms fast autoreclosing and instantaneous autoreclosing have been used in some of the publications included. These terms equal to the more precise term high speed automatic reclosing, which is used throughout the introductory part.

Fault-ride-through

Fault-ride-through means a situation, in which the DG unit surmounts the fault or disturbance without getting disconnected and without losing its stability, thus supporting the network. Fault-ride-through requirements define the most severe disturbance conditions, during which the DG unit must support the network.

Typically these requirements are given for voltage dips as graphs that state the required on-time as a function of voltage dip depth. The origin of fault-ride- through requirements is on the transmission network level, but they are increasingly applied on distribution network level as DG becomes more significant.

(25)

Grid code

Grid codes define the practical requirements for the interconnection of DG unit to the network. Network operators often define their own grid codes according to their own network and needs. Grid codes usually include practical requirements regarding for instance protection devices or voltage control methods. Fault-ride- through requirements are typically included in grid codes.

Loss-of-mains

Loss-of-mains equals to islanding which has been explained earlier in this chapter.

Loss-of-mains refers to the moment during which the connection to the main public system is lost and the island is thus formed.

Network information system

Network information system is a system for managing the data of the distribution network as well as for calculating and planning the network. The system data includes for instance network components, load curves, customer data etc. An essential basis is the combination of technical and economical studies for finding optimal solutions. Network calculation features are integrated to the system. The data structure is often based on databases.

Non-detection zone

An operational area within which the relay is not able to detect the events. In other words, the area within the thresholds of normal operation. In the case of DG unit protection the non-detection zone is often formed by the thresholds of voltage and frequency protection.

Point of common coupling

Point of common coupling is understood to be located at the point where the DG unit connects to the public power system. In the context of protection issues this is located at the same point with the breaker or switch used for disconnecting the DG unit from the public system. Thus multiple DG units may share the point of common coupling if they have a common protection system.

Real-time simulation system

Real-time simulation system is a system operating in a realistic time-scale with a certain time-step. It typically comprises of physical devices and processors for performing the demanding calculation task. The most essential feature is the possibility of using real devices as a part of the simulations due to the realistic time-scale.

(26)

SCADA system

A SCADA (Supervisory control and data acquisition) system is a system for controlling the network equipment. The control is performed through communication channels using various techniques. The control system is typically combined with a data acquisition system that uses the same communication channels for acquiring information from the equipment. This acquired information can be used for state monitoring as well as for statistical purposes.

Stand-by power unit

A stand-by power unit is a generator unit for supplying stand-by power during interruptions. Stand-by power unit is not used in parallel with the network and is often equipped with a two-way switch which physically opens the connection to the public system while running the stand-by unit. The missing possibility of running in parallel with the network is the factor that differentiates stand-by power units from DG. However, DG can act as stand-by power when designed for this purpose.

Zero voltage protection

The term zero voltage protection refers practically to the operation of zero sequence voltage relay, which is in this context located on the utility side of the unit generator transformer. Due to the transformer connection in the studied system, the relay can not be located on the generator side of the transformer. This protection detects the displacement of neutral voltage and disconnects the DG unit from the network.

(27)

2 IMPACTS OF DG ON DISTRIBUTION NETWORK PROTECTION

The present-day philosophy of planning and controlling a radial distribution network is based on the assumption of unidirectional power flow. The power is assumed to be fed to the network from higher voltage levels and distributed further to the customers. Short circuit currents are assumed to behave similarly. At the same time, it becomes assumed that the network does not include significant rotating machines, which should otherwise be taken into account. These assumptions enable relatively simple and economical schemes for achieving a selective operation of protection system with suitable gradation settings. [2], [3]

According to the principles of selectivity [4], only the protective device closest to the fault must operate to disconnect the fault. Thus the rest of the network can be maintained energized.

The propagation of DG on medium voltage and low voltage (LV) levels changes this fundamental basis. The power flows and short circuit currents may even have upstream directions or at least their amplitudes will change due to presence of DG. [5] Thereby the initial schemes applied for instance for feeder protection may become inoperative or less efficient. A typical distribution network has simply not been designed for power generation units with upstream contribution, which may result in problems. [3], [6] The whole distribution system becomes more active as both loading and generation affect the state of the network continuously [7].

One complicating fact is formed by uncertainties related to DG. DG units may remain connected or disconnected depending on different factors. The protection system – as well as the whole network control system – must operate correctly regardless of the state of the DG units. DNO may also be totally unaware of the state of the small DG units.

In addition to protection impacts, DG has other influences on the usage and performance of the distribution system. The most important of these is probably the impact on voltage levels. The modified power flows relate closely to voltage levels. DG can also affect the power quality and overall network reliability. The

(28)

transient stability of the system may also be an issue. These considerations must also be included in the DG interconnection studies, although this thesis is focusing on protection impacts only.

2.1 Protection impacts

“Protective relaying is the term used to signify the science

as well as the operation of protective devices, within a controlled strategy, to maximize service continuity and minimize damage to property and

personnel due to system abnormal behavior.”

P.M.Anderson: Power System Protection [4]

When considering a distribution network with installed DG units, the main concern is the correct co-operation of protection devices during all possible faults.

Network feeders are typically equipped with dedicated protection relays for managing short-circuits and earth faults. The feeding substation is usually equipped with protection for busbar faults, which also acts as a back-up protection for the feeder protection. The DG connection point is equipped with similar relays with dedicated operation characteristics.

The most traditional protection devices of DG connection point are voltage and frequency relays. They are used for detecting abnormalities in the connection point state, which are caused by network faults or other disturbances. Their operation is determined by imbalances of active and reactive power. [8] Their sensitivity and operation times can be adjusted freely. Especially the voltage protection is often set with operation time steps for fast and delayed tripping. The DG connection point is also equipped with overcurrent protection. However, plain overcurrent is not considered a reliable protection factor due to the behaviour of different generator types as explained later. Thus the overcurrent protection often acts as a protection against DG unit’s internal faults and short circuit faults near the DG unit. In many cases it is implemented with LV fuses instead of relays.

Over-/undervoltage, over-/underfrequency and overcurrent functions form the elementary protection of the DG unit. Additionally, the connection point is nowadays usually equipped with loss-of-mains protection, which is often necessary for avoiding situations, in which the DG unit maintains the voltage in the part of the network while the connection to the main system is lost. This is above all a safety issue. A dedicated earth fault relay may also be necessary depending on the network circumstances.

(29)

Coordinating the operation of DG and feeder protection during different faults is definitely a challenging task. A relatively common practice prefers disconnecting the DG unit during the fault to provide the feeder protection with a normal radial fault situation to be cleared. [3] This is highly required in cases in which DG may actually disturb the operation of feeder protection. Even if the DG unit can not disturb the feeder protection, it will need to disconnect immediately after the feeder breaker operation to avoid islanding situation. Thereby it is a good basic rule to disconnect the DG unit in all cases that require action from the feeder it is connected to. It must be kept in mind, that the protection devices protect also the DG unit against unusual situations.

The previous rule leads us to another problem related to the protection coordination. The DG unit has very minor possibilities to differentiate faults requiring action from those that are, for instance, located on other feeders fed by the same substation. If we consider a short-circuit or an earth fault occurring near the substation on the DG feeder and on the adjacent feeder, it is practically impossible to differentiate these situations from the DG unit’s point of view. So far, a quite general solution has been to simply adjust the DG unit to trip during all network faults, thus assuring the safety of the network. In other words, unnecessary trippings of DG units have been allowed in the name of safety. As the amount of DG increases, it will also have more significance during disturbances regarding the frequency and voltages of the network. Fault-ride- through requirements provide that the DG unit must remain stable and thus support the network even during severe disturbances for certain times. However, these fault-ride-through requirements relate only to properties of the DG unit generator operation. The fault-time performance can thus be substantially limited by the PCC protection settings. The question is, whether the connection point protection allows the fault-ride-through when adjusted to assure the network safety?

In the network protection philosophy the safety of the network always comes first, and there is no need to redefine this principle due to the DG. An optimal solution would assure the safety of the network, yet maintaining the DG units in the network as long as possible. This requires coordination methods. Another factor to be kept in mind is the expense of the protection system. Achieving a totally extensive protection system without any gaps may be truly expensive and practically impossible in some cases.

(30)

2.1.1 Impact of generator type

The generator type has an essential impact on the DG unit’s behaviour during faults and disturbances. When considering protection, the generator type is far more crucial than the primary energy resource. Traditionally, the generators are divided into synchronous and induction generators. When applying the power system viewpoint, power electronic converter interface can be considered as a third ‘generator type’ although it is not generator at all. This is as the converter defines the operation of the unit. The converter can be fed by different generators or direct current (DC) sources. Converters are applied especially for exploiting variable power output of primary energy source or for converting DC current to AC. Due to their control possibilities, converters resemble synchronous generators from the network point of view during normal operation conditions. Among the ones listed, new generator techniques are emerging rapidly at the moment. Many of these relate to wind power or other RES (renewable energy source) applications and use power electronic components for control. All new techniques have not been covered in this thesis. However, each one of them can be modelled in the methods presented later.

The most essential difference in fault-time behaviour is the short-circuit current contribution. Traditional synchronous generator is able to feed prolonged short- circuit current. The current contribution is likely to decay after the first cycles, but is often boosted up again by the field forcing. [7] The operation of the field forcing feature is dependent on the generator excitation system. The prolonged contribution can be considered problematic as it may disturb the power system, but the situation is also easy to detect due to the current contribution. At its extreme, the short-circuit currents may constrain the amount of DG units with synchronous generator to be installed in the network. [9], [10] Induction generation may feed an initial short-circuit current that is even greater than the one fed by a synchronous generator. However, the current may decay rapidly during symmetrical short-circuit faults as the excitation is lost. This makes the situation difficult to detect at the PCC of DG. In the case of self-excited induction generator or during unsymmetrical short-circuits, prolonged currents are also possible. [7], [9], [11]

Power electronic converter is probably the most complex component as its behaviour depends strongly on the control system design and on the dimensioning of the converter hardware. The most typical assumption is that the short-circuit current contribution is a current peak with amplitude of nominal current multiplied by certain factor. These factors vary between 1…3 depending on the source and the device considered. [7], [12], [13] It has also been observed, that the short-circuit current contribution of converter-equipped generator is between

(31)

10…20 per cent of that of a similar synchronous generator. [14] In some context, the converter has been proposed to be calculated similarly to the induction generator [13]. Typically the current feed is restricted by the internal protection of the converter bridge. The equipment can also be designed to feed prolonged short- circuit currents. This is not usual at the moment for cost reasons, but it has been proposed to become a requirement. However, no development towards this direction has been reported lately. From the network planning perspective, the lack of general models for the operation of converter devices can be considered problematic.

2.1.2 Impact on short-circuit currents

“In particular, the presence of rotating generators in the distribution network can significantly alter the flow of

fault currents and so needs careful attention.”

Jenkins, Allan, Crossley, Kirschen, Strbac: Embedded Generation [7]

An important impact of DG is caused by the increasing short-circuit currents. As the DG unit contributes to short-circuits, the overall short-circuit currents near the fault point will evidently rise. Problematically high short-circuit currents can also be expected in the vicinity of the DG unit. These currents may result in exceeding the thermal limits of network components. Further, heating impacts or electromagnetic forces may cause damages. [15] As it will be explained later, DG can also delay the operation of feeder protection. Delayed relay operation together with increased short-circuit current is very likely to cause problems. The possibility of failing reclosing explained later is also closely connected to the thermal problems as longer periods with fault on line can be expected.

It is possible to define the DG capacity that can be installed to the network through the network fault current levels. In these general methods the permissible DG capacity is stated as a percentage of the three-phase fault current level at the PCC of DG. These restrictions usually relate to voltage level issues although they are exploiting the fault current levels. [16]

Especially cable connections, transformers or switchgear equipment can be considered problematic with increased fault currents. In [10] it is noticed, that short-circuit current levels may be as high as 90 percent of the switchgear rating prior the interconnection of DG. Also [17] states that devices may often be already at their limits prior the integration of DG. Thus DG installation may require costly upgrades. The capability of the circuit-breaker to interrupt the increased short-circuit current may also become a limiting factor. [18] According to [19], DG can also increase the DC component seen during the breaker opening,

(32)

which should also be taken into account for checking the capability of switchgear.

It must also be noted, that the network reinforcements that are sometimes needed for DG interconnection to keep the voltage levels proper will evidently lead to higher short-circuit currents as well.

In addition to component upgrades, short-circuit current related problems can be handled by other methods. One way of decreasing short-circuit currents is splitting the network in smaller pieces. This has power quality consequences [15]

and is not possible in all networks. It is also possible to dimension the transformer values so that the short-circuit currents are reduced. Use of fault current limiters (FCL) offers a more sophisticated solution. The operation of FCL can be divided in three groups [15]:

1. Fast interrupting devices interrupt the fault current instantly 2. Fault current limiting devices limit the current to a safe level

3. Fault current limiting and interrupting devices are a combination of the previous techniques. Current is limited but also becomes interrupted after a certain time.

Typically the operation is based on increasing the impedance of the FCL significantly during the fault, thereby decreasing the flow of current. The transition is achieved for instance by controlled switches or superconducting materials. The operation of FCL has an impact on voltage dips in the network;

dips can be reduced significantly by applying FCL. [20] FCLs can be installed in substation busbar, outgoing feeders or, in the case of DG, in the PCC. [18], [20]

Limiting the short-circuit current contribution of DG can help to use the present network equipment without damages.

2.1.3 Sensitivity problems

“Sensitivity in protective systems is the ability of the system to identify abnormal condition that exceeds a nominal “pickup” or

detection threshold value and which initiates protective action when the sensed quantities exceed that threshold.”

P.M.Anderson: Power System Protection [4]

Sensitivity problems are possible in cases in which the initial feeder relay settings are not checked as DG is installed in the network. Sensitivity problem means a fault that is not detected at all or is tripped slower than in the initial scheme. It is obvious that this may result in severe safety problems. Additionally, relay operation delays may result in exceeding the thermal limits of network components. It is essential to note, that the overall short-circuit currents will

(33)

increase due to the DG integration, which makes the operation delays more crucial.

The sensitivity problem related to the operation of feeder protection is often called protection blinding. It has been described in detail in publication 2. Briefly, the problem lies in the fact that the short-circuit currents measured by the feeder relay are decreased due to the contribution of DG. [3], [17], [21], [22] Thus the initial relay settings are not evidently valid after installing DG on the feeder. This can be easily proven with normal short-circuit calculations as presented in publication 2.

Blinding takes place in all short-circuit situations in which DG is present.

However, the significance of the phenomenon is strongly dependent on the network, generator type, and the fault location. [3] Figure 2.1 shows a typical situation in which the blinding may take place.

Figure 2.1. Simplified presentation of a situation in which blinding occurs.

The phenomenon occurs in cases in which the DG unit and the substation are feeding a short-circuit fault in parallel. The most important factor is the location of common feed point (CFP) as defined in publication 2. The CFP is defined as the point fed in parallel by the DG unit and the substation that is located closest to the fault. Thereby the location of CFP is fault-specific. Problems are most probable in cases in which a powerful DG unit is located far from the substation.

It must be noted, that very similar problems can be expected in the fuse protection of low voltage networks. [23]

The nature of the problem depends on the type of generator used in the DG unit.

In the case of a powerful synchronous generator, the short-circuit current measured by the feeder relay may become decreased for significant times, resulting thus in total blocking of relay operation at the substation. In the case of network magnetized induction generator, the resulting impact is more likely to be a delay in relay operation. This is as the induction generator’s short-circuit contribution decays rapidly. The operation of permanent magnet induction

(34)

generator might be similar to that of a synchronous generator. DG units equipped with power electronic converters are least likely to cause problems as they usually do not contribute to network short-circuit faults. On the other hand, the behavior of converter applications depends strongly on their design and can thus not be generalized to a high degree. A converter can also be designed to feed a prolonged short-circuit current even at amplitudes greater than the nominal and may thus also be problematic regarding blinding.

Another factor influencing the nature of the sensitivity problems is the operation characteristics of the feeder relays. Specified-time relays are more probable to face total blockings if the tripping threshold is not exceeded any more. The limit distance at which the short-circuit results in delayed or fast operation on the feeder may be moved, which is not problematic in most cases. Dependent-time measuring relays face delayed operation theoretically in all short-circuits. The significance of the delay is strongly case-dependent as described earlier. The dependent-time measuring relay is not likely to face total operation blockings.

However, even the operation delays are problematic in many cases as the thermal limits of the network components as well as the safety requirements are calculated according to certain operation times.

It must be noted that the contribution of DG often leads to small rise of voltage levels in the network. A higher pre-fault voltage results evidently in higher short- circuit currents. In the studies performed in publication 2, this impact was observed to compensate the blinding impact only in the short-circuits closest to the substation. During short-circuits in the tail parts of the feeder, the blinding impact was much more significant. However, the voltage rise phenomenon was observed to reduce the blinding problem in all short-circuits to some extent. If the DG unit participates in voltage control or other measures are applied to keep the voltage rise in minimum, the blinding impact occurs more severely. This issue has been considered in publication 2.

A direct consequence of blinding which may be considered very problematic in many cases is the impact on fault location algorithms. If the measured short- circuit current modified by DG is not taken into account, the fault distances will be miscalculated. Similarly to blinding, the impact is minor between the substation and the DG unit, but increases significantly beyond the CFP point. Due to the DG, faults are calculated to locate further than they actually do. This has been discussed for instance in [24]. In many cases where blinding is not problematic in the form of undetected faults, it may cause problems with fault location. One fact further complicating the situation is that the DG unit may be connected or disconnected depending on the moment, even without the DNO knowing the state of the unit. In the case of bigger units this is typically not a

(35)

problem, whereas small units are often not monitored by the DNO and may thereby cause problems. The aggregate impact of small units can be significant.

Hence definite corrections in location algorithms are not suitable in all situations.

Another area of sensitivity problems are the DG unit’s problems with detecting network faults. In some cases, short-circuit or earth faults occurring on the medium voltage level are difficult to detect from the low voltage side of the DG unit’s transformer. The most problematic issues are loss-of-mains detection and earth fault detection, which are considered later in separate chapters. Problems are also possible during faults that occur far from both the DG unit and the feeding substation.

2.1.4 Selectivity problems

“Selectivity in a protective system refers to the overall design of protective strategy wherein only those protective devices closest to a fault will operate to remove the faulted component.”

P.M.Anderson: Power System Protection [4]

DG-related selectivity issues include two typical problems; the possibility of unnecessarily disconnecting the DG feeder (also called sympathetic tripping) and the possibility of nuisance tripping of DG units. Neither of these events causes an actual safety hazard, but they are of great harm to both producer and network operator. In both cases, IPP suffers a certain amount of energy not produced due to the missing network connection. In the case of sympathetic tripping, the customers of the whole feeder experience a totally unnecessary interruption, which results in reduced reliability from the DNO’s point of view. Also the nuisance tripping of DG results in voltage variations and reduced quality of power supply.

The theory behind sympathetic tripping is extremely simple. A short-circuit fault occurs on another feeder that is fed from the same substation with the DG feeder.

DG contributes to the fault by feeding a short-circuit current upwards towards the substation and further towards the fault. If the feeder relay does not detect the direction of current, it may become nuisance tripped if the current amplitude exceeds the relay threshold. [17], [22], [25] Directional overcurrent relay offers a straightforward solution for this problem. On the other hand, the existing equipment is usually of a non-directional type. Allocating the costs of equipment replacements may be problematic and may constrain the economical viability of DG. It must be noted, that relay upgrading can also offer other benefits apart from the DG-related ones, for instance faster feeder protection. A typical situation during which sympathetic tripping is possible is shown in figure 2.2.

(36)

Figure 2.2. Upstream contribution of DG and the possibility of tripping non- directional relay.

Sympathetic tripping can also be avoided by coordinating the operation times of feeder relays. If the faulted adjacent feeder is tripped faster than the DG feeder, the sympathetic tripping should not occur. Presently a quite common practice is to apply same overcurrent protection characteristics on similar adjacent feeders. This could enable the possibility of sympathetic tripping. It is not always possible to modify the relay operation times in order to avoid sympathetic tripping as other factors may constrain this possibility.

Nuisance tripping of DG unit can occur under similar circumstances as sympathetic tripping. Fault can be located on adjacent feeder, on higher voltage level or at the substation. Due to the protection operation times for these fault locations, short-circuit on adjacent feeder is most likely to cause problems. The deviation of voltage or frequency at the PCC may be great enough to trip the DG unit. [17] Thus the DG protection settings should be assessed for the worst-case faults outside the DG feeder. These faults can be found near the substation but also in areas where the operation of feeder relay with specified-time operation characteristics shifts from one operation mode to another. Sensitivity problems described in the previous chapter may be in conflict with the nuisance tripping problems as they may require opposed protection setting modifications.

Viittaukset

LIITTYVÄT TIEDOSTOT

The main objective of the thesis is to find novel methods for error detection in satellite navigation which are outside of the traditional approach of fault detection and

Thereupon, Fault Detection, Isolation and Restoration (FDIR), better known as Fault Location, Isolation, and Service Restoration (FLISR) [90] can be viewed as one of

tieliikenteen ominaiskulutus vuonna 2008 oli melko lähellä vuoden 1995 ta- soa, mutta sen jälkeen kulutus on taantuman myötä hieman kasvanut (esi- merkiksi vähemmän

Tuulivoimaloiden melun synty, eteneminen ja häiritsevyys [Generation, propaga- tion and annoyance of the noise of wind power plants].. VTT Tiedotteita – Research

Ydinvoimateollisuudessa on aina käytetty alihankkijoita ja urakoitsijoita. Esimerkiksi laitosten rakentamisen aikana suuri osa työstä tehdään urakoitsijoiden, erityisesti

Hä- tähinaukseen kykenevien alusten ja niiden sijoituspaikkojen selvittämi- seksi tulee keskustella myös Itäme- ren ympärysvaltioiden merenkulku- viranomaisten kanssa.. ■

The adaptive IED7 can also provide backup protection in case of transfer trip failure (like the adaptive IED2 does, as explained earlier) in the case of the fault F2 in

When the background network was larger, sinusoidal wave’s amplitude occurred at the shorter feeder length, and when the lowest compensation level was used, the highest